Fig. 1
From: Accurate and instant frequency estimation from noisy sinusoidal waves by deep learning
Schematic of the neural network (NN) model. To prepare the noisy sine wave as the input of the NN model we took 2000 samples from each wave. Each data sample is a node in the input layer of the NN model as is shown here. The NN model has three hidden layers with 2, 2, and 3 neurons in the first, second and third, respectively. The output layer has only one node, which represents the desired frequency. The Nesterov Adam optimizer with a learning rate (lr) of 0.001 was used for this model